Dobrokhotov et al. Clin Trans Med (2018) 7:23 https://doi.org/10.1186/s40169-018-0202-9

REVIEW Open Access Mechanoregulation and pathology of YAP/ TAZ via Hippo and non‑Hippo mechanisms Oleg Dobrokhotov1,2, Mikhail Samsonov3, Masahiro Sokabe2 and Hiroaki Hirata1,2*

Abstract Yes-associated protein (YAP) and its paralog WW domain containing transcription regulator 1 (TAZ) are important regulators of multiple cellular functions such as proliferation, diferentiation, and survival. On the tissue level, YAP/ TAZ are essential for embryonic development, size control and regeneration, while their deregulation leads to carcinogenesis or other diseases. As an underlying principle for YAP/TAZ-mediated regulation of biological func- tions, a growing body of research reveals that YAP/TAZ play a central role in delivering information of mechanical environments surrounding cells to the nucleus transcriptional machinery. In this review, we discuss mechanical cue- dependent regulatory mechanisms for YAP/TAZ functions, as well as their clinical signifcance in progression and treatment. Keywords: YAP, TAZ, Hippo pathway, , Contact inhibition of proliferation, Actomyosin, Cancer progression, Cancer therapy resistance

Introduction gene expression (reviewed in [12]; and [13–16]). While Te transcriptional coactivators YAP and TAZ (Fig. 1) the Hippo pathway is the most widely known regulator have compelled researchers’ attention over the past dec- of YAP/TAZ activity, it is not the sole one. Since the frst ades as important mediators of multiple biological func- identifcation of YAP/TAZ as a mechanotransducer [17], tions. YAP/TAZ play essential roles in development, many groups have shown that YAP/TAZ are regulated homeostasis and regeneration of tissues and organs [1– in response to a number of diferent types of mechanical 6], and their dysfunctions cause several diseases [1, 4, 5, stimuli through Hippo-independent mechanisms. 7]. Both YAP and TAZ are expressed in most of In this review, we frst discuss regulatory mechanisms tissues with some exceptions, for instance TAZ is not of YAP/TAZ focusing on their regulation by mechanical expressed in thymus and peripheral blood leukocytes, cues from cell microenvironments such as adjacent cells while YAP is not expressed in hippocampus and parathy- and the (ECM). We further discuss roid gland [8–10]. the roles of YAP/TAZ in tissue homeostasis and pathol- As a canonical mechanism for YAP/TAZ regulation, ogy with particular interest in their roles in cancer pro- a phosphorylation cascade of the Hippo pathway has gression and resistance against therapeutic treatments. been identifed [11]. Tere are two major steps in Hippo Importantly, while YAP and TAZ have a lot of similari- pathway-mediated regulation of YAP/TAZ: YAP/TAZ ties in their structures, regulations and functions, they phosphorylation and nuclear translocation of YAP/TAZ. are not identical (Fig. 1; [8, 18, 19]). Terefore, in this Nuclear YAP/TAZ mainly interact with transcription review we use the term “YAP/TAZ” when there is evi- factors of the TEA domain (TEAD) family members, as dence that YAP and TAZ share the described functions well as several other transcriptional factors to regulate or regulations, whilst we specify the single protein name when evidence is available only for either YAP or TAZ.

*Correspondence: [email protected]‑u.ac.jp 2 Laboratory, Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa‑ku, Nagoya 466‑8550, Japan Full list of author information is available at the end of the article

© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creat​iveco​mmons​.org/licen​ses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 2 of 14

S127 N C

YAP P-rich TEAD binding WW WW CC TAD SH3-BM PDZ-BM

S89 N C

TAZ TEAD binding WW CC TAD PDZ-BM Fig. 1 Domain structures of YAP and TAZ. YAP is a 65-kDa protein containing a -rich region (P-rich) in the N-terminal region and a PDZ-binding motif (PDZ-BM) in the C-terminal region that are separated by two WW domains, a TEAD binding domain, a Src homology domain 3 binding motif (SH3-BM), and a coiled–coil domain (CC) within the transactivation domain (TAD). There are 8 isoforms of YAP exist which difer by loss of one WW domain and alterations in TAD. TAZ, a 43-kDa paralog of YAP, has similar domain organization but lacks the proline-rich region, the SH3-BM and one WW domain. YAP S127 and TAZ S89 are the main phosphorylation targets of LATS1/2. Upon phosphorylation YAP/TAZ binds with 14-3-3 and thus are sequestered in the cytoplasm

Canonical YAP/TAZ regulation via Hippo pathway cytoplasm [28, 31, 32]. Second, α-catenin does not always Te core Hippo pathway starts with MST1/2—STE20 solely localize at AJs but shows signifcant cytoplasmic family protein , which phosphorylate hydrophobic attendance [27, 33]. Last, α-catenin shows a protec- motifs of LATS1/2 [20]. Phosphorylated, i.e. activated, tive efect against PP2A-mediated dephosphorylation LATS1/2 then phosphorylate serine residues in YAP/ of YAP in an in vitro dephosphorylation assay without TAZ [11, 21], leading to cytoplasmic retention of YAP/ membrane supports [27], indicating its ability to act as a TAZ [21] (Fig. 2). In addition to direct phosphorylation soluble form. As such, the cytoplasmic complex forma- of LATS1/2 by MST1/2, MTS1/2-phosphorylated SAV1 tion with α-catenin would ensure retention of YAP in the and MOB1A/B facilitate phosphorylation of LATS1/2 cytoplasm. [22–24]. While the role of MOB1A/B is not fully under- Another mechanism for inhibiting nuclear localization stood (Meng et al. proposed a model in which MOB1A/B of YAP/TAZ is phosphorylation of YAP/TAZ by casein act as scafold proteins, but it has not been experimen- K1δ/ε following the LATS-dependent phosphorylation. tally justifed yet [25]), SAV1 binds to MST1/2, targeting Such polyphosphorylation induces YAP/TAZ ubiquitina- the whole complex to the plasma membrane [26]. For tion by the SCF E3 ubiquitin ligase and subsequent deg- quite a long time, MST1/2 have been deemed to be the radation [25]. primary kinases that phosphorylate LATS1/2. However, To date, several upstream regulators of the Hippo path- it has been shown that double-depletion of MST1/2 does way have been identifed: TAO1/2/3 kinases that phos- not alter YAP phosphorylation [27, 28]. Indeed, a recent phorylate the activation loop of MST1/2 [25, 34, 35], study has shown that MAP4K family kinases act in paral- G-protein coupled receptors [36], Kibra [37], PTPN14 lel to MST1/2 to activate LATS1/2 [29]. [37], AMPK in response to energy stress [38–40], and Upon phosphorylation by LATS1/2, YAP/TAZ inter- Ras-mitogen activated protein (MAPK) signalling act with 14-3-3, which sequesters YAP/TAZ from trans- [36, 41]. Moreover, while cells sufer various mechani- location into the nucleus [21]. α-Catenin, a well-known cal stimuli of diferent origins, including cell–ECM and adaptor protein localizing at -mediated cell– cell–cell contacts, ECM stifness, fuid shear, cell geom- cell junctions, stabilizes the YAP complex with 14-3-3, etry, internal actomyosin tension and so on [1, 25, 42], thereby inhibiting YAP dephosphorylation by protein these mechanical stimuli also strongly afect YAP/TAZ phosphatase 2A (PP2A) [27, 30]. It has been suggested regulation. that α-catenin acts at adherens junctions (AJ) to sta- bilize the YAP complex [27]. However, in our opinion, Impact of cell–ECM interaction on YAP/TAZ α-catenin may interact with the YAP-14-3-3 complex regulation both at and outside AJs. While this idea requires further Cells adhere to ECM through the macromolecular adhe- justifcation, there are several observations supporting sion complex called focal adhesion (FA) that links the this idea. First, exclusion of YAP from the nucleus usu- actin to ECM [43–45]. While cells sense ally does not lead to its plasma membrane recruitment, stifness of ECM and change their spreading and migra- rather it more or less evenly distributes throughout the tion in response to ECM stifness [17, 31, 46–52], YAP/ Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 3 of 14

Stiff and Fibronectin-rich ECM

Fig. 2 Overview of signaling cascades for YAP/TAZ regulation. Hippo pathway- (blue), FA- (pink) and AJ-mediated (yellow) regulations of YAP/TAZ are shown. See the main text for details

TAZ activity is regulated by ECM stifness and cell and Src–Rac1–PAK axes form a single cascade or act in spreading. Spread cells and cells grown on stif ECM parallel. show elevated YAP/TAZ activity which correlates with On the other hand, Dupont et al. reported that their nuclear localization [17, 31, 48, 53]. mechanical cue-induced regulation of YAP in human Mechanical environments characterized by cell mor- mammary epithelial cells and human mesenchymal stem phology and the cell–ECM contact area regulate YAP cells is independent of the Hippo pathway [17], which nuclear localization via afecting LATS1/2-dependent was later confrmed by other groups [31, 48, 58]. How- YAP phosphorylation [53]. As an underlying molecu- ever, for a long time, the mechanism that provides con- lar mechanism, involvement of signalling has sistent explanation for the mechanical regulation of YAP/ been revealed (Figs. 2 and 3). Stif or fbronectin-rich TAZ has been unsolved. ECM activates the β1-integrin–FAK–Src–PI3K–PDK1 It took some time to fnd it, but recently Elosegui- pathway, which inhibits the LATS1/2 activity and thus Artola et al. have reported that the nucleus itself plays facilitates YAP nuclear translocation and transcrip- a role as a mechanotransducer for YAP regulation [59]. tional activation [54–56]. Te Src–Rac1–PAK path- Tey have shown that the actin cytoskeleton, in particu- way also inhibits the YAP phosphorylation by LATS1/2 lar actomyosin fbers running from FAs to the apical sur- downstream of β1-integrin [57]. PAK activation induces face of the nucleus [60], afords mechanical connection phosphorylation of , which abrogates its scaf- between FAs and the Linker of the Nucleoskeleton and fold function for YAP and LATS1/2, and thereby atten- Cytoskeleton (LINC) complex of the nuclear envelope uates YAP phosphorylation by LATS1/2 [57] (Fig. 2). It on stif substrates, but not on soft ones. Contractile force is currently unknown whether FAK–Src–PI3K–PDK1 generated by actomyosin activity fattens the nucleus and Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 4 of 14

opens up nuclear pores. Such opening increases the rate YAP is facilitated, it does not fully explain why signif- of YAP/TAZ import but not afecting the export rate cant changes in the pore shape and size do not afect the (Fig. 4a). Tis mechanism might also explain cell area- export rate. Although we do not have a reasonable model and cell shape-dependent regulation of YAP/TAZ locali- that might explain all observations, it is necessary to take zation [59]. notice that the mechanism of nuclear transport is rather As an attempt to explain the diferent infuences of complex. In the case of YAP, its export from the nucleus mechanical tension on import/export rates, the following would likely depend on the concentration of free nuclear model was proposed. Te inner lumen of nuclear pores YAP (i.e., YAP that is not trapped by TEAD [63] or other comprises a disorganized meshwork of fexible FG-nups transcription factors in the nucleus), as well as availability that contain phenylalanine–glycine (FG) repeats [61, of Merlin in the nucleus, wherein nuclear export signals 62]. Repulsive interaction of proteins with FG repeats of Merlin are suggested to mediate YAP nuclear export leads to protein unfolding and thus facilitates protein [64]. passage through the nuclear pore. Furthermore, tension Nuclear localization of proteins is mediated by two in the cytoskeleton-nucleus system not only increases distinct mechanisms: passive and active. Passive difu- the nuclear pore size but also increases the curvature of sion through nuclear pores is suitable for proteins smaller the lateral part of the nuclear membrane. Tis leads to than ~ 40 to 50 kDa (note, theoretical MWs of YAP and showing up the inner surface of the nuclear pore to the TAZ are ~ 65 and ~ 43 kDa, respectively [18]). Te sec- cytoplasm, thus promoting FG-nups-mediated translo- ond mechanism is the energy-dependent process that cation of cytoplasmic proteins into the nucleus (Fig. 4a). involves transporting proteins (importins and exportins). While their model could explain how nuclear import of It is known that YAP/TAZ nuclear shuttling follows the active mechanism [17, 59, 64, 65]. Importantly, an increase in nuclear pore diameter under high tension in the actin cytoskeleton does not switch the mechanism of YAP/TAZ nuclear transport from active to passive; nucleus–cytoplasm translocation of YAP/TAZ under high tensile conditions (e.g., on stif ECM) still depends on importin and exportin activities [59]. Tus at least two distinct mechanisms underlie cell– ECM interaction-mediated activation of YAP/TAZ: inhi- bition of LATS1/2-mediated phosphorylation of YAP/ TAZ, and an increase in YAP/TAZ import through open- ing of the nuclear pores, which might synergistically reg- ulate YAP/TAZ transcriptional activity (Figs. 3 and 4a).

Cell–cell contact‑mediated regulation of YAP/TAZ Sensing neighboring cells is essential for tissue integrity. Cell–cell contact-induced arrest of cell growth, termed contact inhibition of proliferation (hereafter contact inhi- bition, CIP), underlies homeostatic control of cell den- sity and organ size [30, 66]. Involvement of YAP/TAZ in cell contact inhibition and tissue growth control has been revealed over the decade [21, 30, 67]. However, the detailed mechanism underlying YAP/TAZ-mediated CIP is still poorly understood. At a low cell density condition the phosphorylation level of YAP is low and it predominantly localizes in Fig. 3 A hypothetical model for YAP/TAZ regulation by tension at FAs. a Low tension FA with small actomyosin force and/or soft ECM. b the nucleus, while at a dense condition YAP is highly High tension FA with large actomyosin force and stif ECM. FA tension phosphorylated and sequestered in the cytoplasm [17, induces sequential phosphorylation (P) of FAK, Src and p130Cas (Cas). 21, 64]. Under a high cell density condition, LATS1/2 Phosphorylated Src activates the PI3K–PDK2 pathway, whilst the are phosphorylated and activated by MST1/2 and phosphorylated p130Cas activates the Rac1–PAK pathway. Activation MAP4K1/2/3/4/6/7 [17, 21, 28, 29], and double-deple- of these pathways inhibits LATS-mediated phosphorylation of YAP, facilitating nuclear translocation of YAP tion of LATS1/2 abrogates cell density-dependent phos- phorylation of YAP [29], indicating that LATS1/2 are Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 5 of 14

cytoplasmic side, β-catenin binds to the cytoplasmic domain of cadherin, as well as interacts with the actin binding protein α-catenin. It is well established that E-cadherin–β-catenin–α-catenin-actin linkage provides a major mechanical connection of the actin cytoskeleton to E-cadherin, which transmits actomyosin-generated ten- sile force to E-cadherin at AJs [72]. Stability and function of AJs depend on trans interaction of the cadherin extra- cellular domains and actomyosin-generated mechani- cal tension [71]. A number of other proteins have been shown to localize at AJs. E-cadherin reportedly mediates CIP in a Hippo path- way-dependent manner [28]. While E-cadherin homo- philic ligation in MCF-7 and MCF-10 cells leads to inhibition of , this process depends on other AJs components, including α-catenin, β-catenin, NHERF ­(Na+/H+ exchanger regulatory factor) and Mer- lin, as well as the Hippo pathway components Kibra and LATS1/2. Interestingly, MST1/2 are not required for this process. Involvement of E-cadherin in the cell density-depend- ent regulation of YAP has been further supported by the Fig. 4 Hypothetical models for actomyosin-dependent regulation of fndings that MDA-MB-231 cells, which do not express YAP nuclear localization in sparse and confuent cells. a Sparse cells E-cadherin, show nuclear localization of YAP even develop FAs that are connected to actomyosin stress fbers. Tensile force exertion from stress fbers to FAs activates the FAK–Src signal, under the high cell density condition, and that exoge- which inhibits LATS1/2-mediated phosphorylation of YAP, thereby nous expression of E-cadherin in these cells redistributes facilitating nuclear translocation of YAP. In addition, contraction of YAP from the nucleus to the cytoplasm [28]. It is note- stress fbers connecting FAs and the apical surface of the nucleus worthy that expression of the E-cadherin mutant, which (sometimes called ‘actin cap’) fattens the nucleus, increases the cannot bind to the catenin complex but is still able to curvature of the lateral part of the nuclear membrane, and thereby enlarges the diameter of the cytoplasmic side of the nuclear pore in mediate cell–cell adhesions, does not cause relocaliza- this nuclear membrane part. Such ‘asymmetric opening’ of nuclear tion of YAP into the cytoplasm. By contrast, expression pores may preferentially promote nuclear import, rather than of an E-cadherin/α-catenin chimeric protein that can export, of YAP. b Confuent cells are poor in FAs and stress fbers, directly bind to the actin cytoskeleton induces cytoplas- but instead develop AJs and actomyosin fbers associated with mic sequestration of YAP, even more efciently than AJs. Actomyosin-based tensile force at AJs induces translocation of Merlin from AJs to the nucleus, wherein Merlin forms a complex with that of wild-type E-cadherin. Tese results suggest that YAP. With the aid of nuclear export signals of Merlin, the Merlin-YAP connection of E-cadherin to the actin cytoskeleton con- complex is then exported from the nucleus. Thus the actomyosin tributes to sequestering YAP in the cytoplasm at a dense activity has opposing efects on the YAP distribution between condition. Consistently, depletion of either β-catenin or sparse and confuent cells; the actomyosin activity promotes nuclear α-catenin induces nuclear localization of YAP in dense localization of YAP in sparse cells, but attenuates it in confuent cells. See detailed discussion in the main text cell cultures, as well as in mouse skin [28, 30]. Anyway, there is no doubt that E-cadherin is essential for cell den- sity-dependent subcellular localization of YAP in epithe- lial cells. responsible for phosphorylation of YAP/TAZ in response Recently, two independent groups have revealed that to cell density. actomyosin-based tension at AJs suppresses nuclear Several studies have revealed important roles of AJ localization of YAP/TAZ in high density epithelial cells, components in YAP regulation [27, 30, 32, 68, 69]. AJ thereby arresting cell growth [32, 64] (Figs. 4b and 5). is a macromolecular complex that mediates adhesion Tus inhibition of actomyosin or disconnection of the between cells and is composed of transmembrane cad- actin cytoskeleton to AJs (by depletion of α-catenin or herin family proteins (E-cadherin in the case of epithelia) β-catenin) causes nuclear localization of YAP in high and their associated catenins [70, 71]. Te extracellular density cells as well as in the mouse skin [28, 30, 32, domain of cadherin is responsible for homophilic inter- 64]. Application of external tensile force to E-cadherin actions of cadherin between neighboring cells. On the using E-cadherin-coated magnetic beads, in turn, arrests Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 6 of 14

YAP-driven proliferation of actomyosin-inhibited cells, while E-cadherin homophilic ligation per se promotes cell proliferation under actomyosin inhibition [32]. Tus, considering importance of tension at AJs, we suggest that not E-cadherin by itself but mature AJs connected to actomyosin network would be required for cell density- dependent YAP regulation and CIP. As a molecular mechanism for the AJ-tension-depend- ent regulation of YAP/TAZ localization, Furukawa et al. have shown that YAP/TAZ are excluded from the nucleus dependently on Merlin and its nuclear export signals (NESs) [64]. While Merlin associates with AJs, actomy- osin-based tension at AJs releases Merlin from AJs and allows it to enter the nucleus. Nuclear Merlin binds YAP/ TAZ and facilitates cytoplasmic translocation of YAP/ TAZ with the aid of Merlin NESs (Fig. 4b). Of note, Merlin acts in a Hippo-independent manner; silencing of Merlin expression does not afect phosphorylation of LATS and YAP. Detailed mechanism for tension-dependent release of Merlin from AJs remains to be revealed. However, it is known that Merlin localizes at AJs through the bind- ing with α-catenin [73]. α-Catenin is currently the solely revealed mechanosensor at AJs that alters its molecular Fig. 5 A hypothetical model for YAP/TAZ regulation by tension at AJs. interaction through the force-induced conformational a Low tension AJ with small actomyosin force. Under this condition, change [74, 75]. Tis leads to an intriguing possibility that Merlin localizes to AJs through its binding with α-catenin. b High tension-induced conformational changes of α-catenin tension AJ with large actomyosin force. AJ tension causes dissociation of Merlin from AJs, which may be associated with a force-induced may be involved in dissociation of Merlin from α-catenin conformational change of α-catenin. Released Merlin enters the at AJs, which needs to be tested in future studies (Fig. 5). nucleus, binds nuclear YAP, and then exports it from the nucleus with Even though Furukawa et al. have revealed that Mer- the aid of NESs in Merlin lin deactivates YAP/TAZ by the Hippo-independ- ent mechanism, Merlin can regulate YAP/TAZ also through the Hippo pathway. Tus Merlin functions as a of Merlin-defcient liver progenitor cells was shown to be scafold protein that binds to LATS1/2 and targets them independent of YAP [77]. Tus the overall mechanism of to the plasma membrane in Drosophila and mammalian Merlin-dependent regulation of cell growth is still unclear, HEK293 cells [26]. Merlin-dependent LATS recruit- and further studies are needed to unveil it. ment to the membrane facilitates LATS1/2 phospho- It is known that Merlin localizes to AJs through its rylation by MST1/2 and MAP4Ks, followed by YAP/ binding to α-catenin, and silencing α-catenin expres- TAZ phosphorylation by LATS1/2 [26, 29], which sup- sion leads to delocalization of Merlin from AJs [73]. Even ports the Hippo-dependent regulation model. Hence, though an increase in non-AJ Merlin would accelerate we could speculate that Merlin regulates YAP/TAZ by export of YAP from the nucleus [64], α-catenin deple- multiple ways and diferent mechanisms may prevail in tion, instead, leads to an increase in nuclear YAP [32, 64]. diferent cells and under diferent conditions (Fig. 2). Tis suggests that α-catenin regulates YAP localization Merlin-dependent regulation of intracellular localiza- not only through Merlin but also via other mechanism(s). tion of YAP/TAZ has been confrmed also in vivo. Con- As discussed in the former section (“Canonical YAP/TAZ ditional knockout of Merlin in mouse liver results in regulation via Hippo pathway”), cytosolic α-catenin may signifcant liver expansion which coincides with decreased prohibit nuclear translocation of YAP by forming a com- phosphorylation and increased nuclear accumulation plex with YAP. Similarly, β-catenin also forms a cytosolic of YAP [76]. Both, impairment of Merlin-driven nuclear complex with YAP as the β-catenin destruction complex, export of YAP/TAZ and increased YAP/TAZ nuclear which contributes to sequestering YAP in the cytoplasm import due to a decrease in YAP/TAZ phosphorylation [78]. Tese results suggest that AJ components act not would contribute to enhanced nuclear accumulation of only as AJ-associated forms but also as soluble forms in YAP/TAZ under Merlin knockout. Notably, overgrowth the regulation of YAP/TAZ localization. Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 7 of 14

As described above, tension at AJs leads to sequestra- hand, while Src directly phosphorylates three tyrosine tion of YAP/TAZ from the nucleus in high density cells residues in the transcription activation domain of YAP, [32, 64]. On the other hand, AJ tension under a lower cell α-catenin antagonizes ECM-induced activation of the β4 density condition, in which cells form AJs but still pro- integrin–Src pathway, leading to inactivation of YAP [86] liferate, may have opposing efects on YAP localization. (Fig. 2). Under such condition, tension at AJs activates vinculin to recruit the LIM protein TRIP6 to AJs, which sequesters LATS1/2 at AJs and thereby inhibits LATS1/2 activation YAP regulates mechanical properties of tissues by MST1/2 and MAP4Ks, resulting in a decrease in YAP In the above sections, we have discussed how mechanical phosphorylation [79]. Tus AJ tension potentially regu- cues from ECM and actomyosin tension regulate YAP/ lates YAP/TAZ both positively and negatively depending TAZ transcriptional activity. Interestingly, reverse regu- on circumstances, even though the mechanism by which lations also exist. It has been recently revealed that YAP it takes opposing tasks is currently unknown and needs increases actomyosin-based tension in tissues (mouse to be revealed in future studies. lung and fsh embryo) by upregulating the RhoA-acto- axis [87, 88]. YAP promotes expression of the Diferential regulations of YAP/TAZ by cell–cell Rho guanine exchange factor (RhoGEF) ARHGEF17 that and cell–ECM interactions activates RhoA [87]. Expression of several regulators and Te fnding that actomyosin-based tension at AJs causes components of the actomyosin cytoskeleton, including cytoplasmic sequestration of YAP/TAZ in high den- myosin IIB, myosin regulatory light chain 2 and flamin sity epithelial cells may sound contradictory to previous A, is also enhanced by YAP either transcriptionally or studies showing that actomyosin contractility facilitates non-transcriptionally [87, 89, 90]. Given that actomyo- YAP/TAZ nuclear translocation in many types of cells sin tension induces YAP activation in many types of cells [17, 31, 59]. However dominant types of the actomyo- (see previous sections), YAP and RhoA signals are likely sin cytoskeleton are diferent between confuent epithe- to be mutually regulated via a feedback loop. lial cells and the sub-confuent cells used in the previous On the other hand, YAP reportedly upregulates expres- studies; while sub-confuent cells show prominent stress sion of the Rho GTPase activating proteins (RhoGAP) fbers connecting to FAs, confuent epithelial cells are ARHGAP18/29 which deactivates RhoA [88, 91]. Tus poor in stress fbers but develop actomyosin cables that expression of both a RhoA activator (ARHGEF17) and associate with AJs [32]. We speculate that actomyosin- RhoA inhibitors (ARHGAP18/29) is promoted by YAP. It based tension might have opposite efects on YAP/TAZ is currently unclear how opposing RhoA regulations by localization depending on the cellular context. Tus ten- YAP through diferent RhoA regulators are spatiotempo- sion at FAs induces FAK phosphorylation [80], which rally coordinated to fne-tune actomyosin systems. causes YAP/TAZ activation [54, 55], as well as “opens” Te mechanical property of ECM is also regulated by nuclear pores for YAP/TAZ nuclear import [59] (Figs. 3 YAP/TAZ. In cancer tissues, activation of YAP/TAZ in and 4a). On the other hand, actomyosin tension at AJs cancer-associated fbroblasts (CAFs) promotes expres- deactivates YAP/TAZ (Figs. 4b and 5). Even though acto- sion of the ECM proteins such as laminin and fbronec- myosin inhibition potentially eliminates both efects, it tin, and induces stifening of ECM, which contributes may cause YAP deactivation in sub-confuent cells that to maintenance of cancer stem cells and CAFs in cancer have developed FAs but not AJs. By contrast, actomyosin environments [89, 92–94]. Tis YAP/TAZ-mediated reg- inhibition in confuent epithelial cells would mainly abro- ulation of ECM would contribute to cancer progression gate the AJ-dependent suppressive efect on YAP, leading and therapy resistance, as discussed below. to activation of YAP. Detailed mechanisms of such diferential regula- YAP/TAZ in tissue homeostasis tion is currently unknown. However, based on several Te Hippo pathway was originally discovered in Dros- recent reports, we could suggest that diferential roles of ophila. Te efect of the YAP/TAZ on the organ size con- α-catenin in YAP/TAZ regulation might be involved. On trol is preserved among species from fies to one hand, phosphorylation of the Ajuba family protein [5]. However, sensitivity to YAP/TAZ dysregulation var- LIMD1 by JNK induces the complex formation of LIMD1 ies among diferent organs/tissues. Tissue specifc knock- with LATS1/2 [81–84], and switch of α-catenin to open out of YAP in virgin mammary gland and intestine does conformation under actomyosin-generated tension at not result in any disorders in tissue size and structure AJs leads to recruitment of LIMD1–LATS1/2 complex to [95, 96], while double conditional skin-specifc knock- AJs and thereby inhibits the kinase activity of LATS1/2, out of YAP and TAZ results in dramatic loss of hair [56]. resulting in activation on YAP/TAZ [85]. On the other TAZ deletion causes the polycystic kidney disease [97], Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 8 of 14

whereas YAP deletion leads to hypoplastic kidneys [98] might not be sufcient for cancer initiation in some tis- and a bile duct defect in liver, without general reduction sues [5, 95, 130–132]. Importantly, mutations in Hippo in organ size [5]. It is noteworthy that in most studies, pathway components are rare in human tumors [133]. in vivo functions of YAP/TAZ have been examined by Mutations in Merlin and LATS1/2 may occur only in spe- knocking-out either YAP or TAZ. Since their functions cifc tumor types, such as mesothelioma, schwannomas, are largely overlapping each other, it is possible that the and meningiomas, but are not observed in most tumor efect of TAZ knockout may be compensated by YAP and types displaying elevated YAP/TAZ activity [107]. Based vice versa. on these results, Zanconato et al. reasonably suggest and While knockout of YAP/TAZ does not give rise to argue that Hippo signaling is not a sole and dominant apparent in some organs in a steady state, mechanism for regulating YAP/TAZ activity in human they appear to play signifcant roles during growth phases tumors [107]. of tissues. For example, hyperactivation of YAP leads Tese considerations raise following questions: to a failure in terminal diferentiation of the mammary how YAP/TAZ are activated in cancer and what is the gland during pregnancy [95]. YAP activity is induced sequence of events downstream of YAP/TAZ activation? upon hepatectomy, which is most likely critical for liver To date mechanisms for activation and action of YAP/ regeneration [99, 100]. Furthermore, loss of YAP causes a TAZ in cancer are poorly understood. However, some reduction of the regenerative ability of intestine [96]. cues have been revealed in recent studies. YAP/TAZ are highly active in progenitor and stem As was mentioned above, YAP and β-catenin syner- cells in multiple tissues and infuence cell fate. Active gize in cell cycle progression. Importantly, YAP overex- nuclear YAP/TAZ promote proliferation and self-renewal pression in combination with β-catenin in mouse liver of embryonic stem cells (ESC) [101–103] and basal leads to the development of hepatoblastoma [134], and keratinocytes [27, 48, 103], as well as osteocyte difer- YAP is required for cell survival in β-catenin-driven entiation of mesenchymal stem cells (MSC) [104, 105]. colon [119]. In both cases, YAP forms a com- On the other hand, cytoplasmic retention of YAP/TAZ plex with β-catenin. Rosenbluh et al. have found that the results in neuronal diferentiation of ESC [101, 102], adi- β-catenin–YAP–TBX5 complex formation drives expres- pocyte diferentiation of MSC [104, 106], and terminal sion of antiapoptotic genes, including BCL2L1 and BIRC5 diferentiation of keratinocytes [48, 56]. [119]. Consistent with this, in BRAF-mutated non-small cell lung cancer (NSCLC) and melanoma, YAP-depend- YAP/TAZ in cancer ent expression of BCL2L1 allows cancer cells to escape Dysregulation of YAP and TAZ could be observed in from under BRAF- or MEK-targeting therapy many tumors of diferent origins, including prostate can- [127], even though it has not been examined whether cer, cervical squamous cell carcinomas, meningioma, β-catenin is involved in this process. squamous cell carcinoma of skin, malignant mesothe- Another role of YAP/TAZ in therapy resistance of can- lioma, acute myeloid leukemia, and others (comprehen- cer was found in studies of BRAF V600E gain-of-function sively reviewed in [107]). Curiously, cancer-associated, mutant melanomas. YAP/TAZ-dependent actin cytoskel- activating mutations of YAP/TAZ have not been found, eton remodeling and cell spreading promote an acquisi- and YAP1 and TAZ genes are rarely amplifed in cancer tion of a resistant against the treatment with cells [108]. Instead, YAP/TAZ are often overexpressed the BRAF inhibitor PLX4032 (vemurafenib) in melanoma and highly accumulated in the nuclei in cancer cells [107, cells in vitro [128]. Even though cytoskeleton remodeling 109]. Tis phenotype is preserved in many cancer tissues per se may not be sufcient for therapy resistance, it is [110–113]. Apparently, YAP/TAZ overexpression pro- possible, as discussed in the previous sections, that cell motes transcription of YAP/TAZ target genes [114, 115]. spreading would facilitate nuclear entry of YAP (and, High YAP/TAZ activity drives proliferation, invasion most likely, other transcription factors and co-activa- and metastases of cancer cells [4, 116–119], and is often tors), and thus might promote cell cycle progression and associated with poor prognosis [120, 121], therapy resist- inhibit apoptosis in cancer cells. ance, including resistance to chemotherapeutic drugs YAP plays a key role also in acquiring EGFR-tyrosine [118, 121–125], radiation [124, 126] and molecularly tar- kinase inhibitor-resistant as well as cetuximab-resistant geted therapies [127, 128], and relapse [107] of cancers. properties of cancers [135–139]. Although details of the In hematological cancers, however, low YAP1 expression underlying mechanism remain unknown, several tran- has been shown to be predictive of a poor outcome of scriptional targets of YAP, including AXL [136, 140], treatments [129]. ERBB3 [141], PD-L1 [139], are involved in acquisition of YAP/TAZ activity is usually essential for tumor pro- these chemo-resistant phenotypes of cancers. gression, even though activation of YAP/TAZ solely Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 9 of 14

Cell invasion and metastasis are typical properties of Tus, in a wide range of cancers, YAP might be regulated malignant cancers. Notably, these activities of cancer by the crosstalk between CAFs, ECM and cancer cells, as cells are also potentially promoted by YAP. Recently, Qiao described in “YAP regulated mechanical properties of tis- et al. have proposed a mechanism for YAP-mediated sues”; YAP activity in CAFs might induce ECM remode- cancer metastasis [91]. First, the high expression level of ling that in turn would induce YAP nuclear translocation YAP in cancer cells promotes epithelial–mesenchymal and phosphorylation by Src family kinases in cancer cells transition (EMT) via reprogramming of gene expression (Fig. 6). Although, the initial trigger that starts up this to antagonize AJ assembly [90, 94], which leads to loss of self-inducing program stays unknown, several studies cell–cell adhesions and apical-basal polarity, along with support an idea that soluble factors secreted by cancer acquisition of mesenchymal motility. Importantly, over- cells might act as such a trigger [36, 89, 147, 149]. expressing YAP in normal cells antagonizes maturation of Considering involvement of YAP and TAZ in cancer AJs, but does not induce EMT [90], suggesting that YAP development, progression and therapy resistance, these cooperates with some other cancer-associated factors to molecules and their upstream regulators would serve as provoke EMT in cancer cells [94]. Next, YAP-driven actin potent targets for cancer therapy [18, 109, 123]. Kinases cytoskeleton remodeling decreases the rigidity of cancer are usually deemed to be the best targets for small mol- cells [91]. Soft cancer cells are easy to squeeze through ecule therapeutics. However, most kinases in the Hippo ECM and extravasate through vasculature to achieve dis- pathway are tumor suppressors, and restoring functions tal metastasis via the circulation [142]. Furthermore, YAP of tumor-suppressor kinases in cancers is a challenging enhances the membrane-cytoskeletal integrity, increas- task [109]. Alternatively, Src, which directly phospho- ing cell viability when cells travel through narrow spaces rylates the YAP activation domain [86], may provide a such as capillary vessels during metastasis [91, 143]. potential target. Te question regarding the mechanism of YAP/TAZ Another approach for YAP/TAZ-targeted cancer activation in cancer might be even more complex, con- therapy is modulating the actin cytoskeleton. Recent sidering potential existence of cell–cell and cell–ECM fndings that actomyosin contractility in epithelial crosstalks in cancer environments [144, 145]. Although sheets induces nuclear exclusion of YAP and thereby melanoma cells in vitro undergo cell death upon the inhibits cell proliferation [32, 64] (see “Cell–cell con- treatment with the BRAF inhibitor PLX4720, these tact-mediated regulation of YAP/TAZ” section) ofer cells are resistant to PLX4720 treatment in vivo [146]. a potential strategy for suppression of YAP-driven cell In vivo treatment of melanoma with PLX4720 acti- proliferation by exogenous activation of actomyosin. vates melanoma-associated fbroblasts (MAFs), leading However, it is important to notice that global induc- to ECM remodeling [146]. Changes in ECM composi- tion of the actin cytoskeleton contractility would have tion and stifness cooperatively provide “safe havens” for harmful efects such as hypertension and asthma. melanoma cells, making them insensitive to PLX4720 Terefore, proper drug dosage and targeted drug deliv- treatment [146]. Te efect of MAF-dependent ECM ery would be critical to implement cancer therapy that remodeling can be recapitulated in vitro. When co- targets actomyosin activity. cultured in gel, MAFs isolated from PLX4720- Inhibition of YAP/TAZ interaction with its transcrip- resistant cancer tissues make melanoma cells resistant tional partners is the third potential strategy. From to PLX4720 through formation of dense collagen fbers screening a library of FDA-approved drugs targeting the [146]. Importantly, melanoma cells re-isolated from the YAP–TEAD complex, Liu-Chittenden et al. identifed co-culture system remains sensitive to PLX4720. Sig- verteporfn (VP) as one of the top hits [150]. While VP nifcance of the ECM property has been further demon- is already used clinically for treatment of a non-cancer strated by the observation that melanoma cells become disease (macular degeneration), results in vitro and in PLX4720-resistant when solely cultured on stif and mouse cancer models indicate VP as a promising com- fbronectin-rich substrates in vitro [146]. ponent of synthetically lethal strategies in melanoma and Generation of the fbronectin-rich stif matrix induces NSCLC treatments in combination with vemurafenib and reorganization of integrin β1 into focal adhesions, ele- erlotinib, respectively [141, 151]. A peptide derived from vating FAK phosphorylation [54, 146]. As discussed in vestigial-like protein 4 (VGLL4), a protein that competes the previous sections, it is conceivable that stif ECM with YAP in binding to TEAD and as such inhibits YAP and phosphorylated FAK activate YAP/TAZ either via function, also shows a therapeutic potential for cancers the Hippo pathway or independently of it. Similar ECM [152, 153]. In addition, fufenamic acid was recently iden- remodeling by mammary adenoma- and carcinoma-asso- tifed as another inhibitor of YAP–TEAD-dependent ciated fbroblasts has been observed [89, 147], and ECM transcription [154]. In contrast to the case of the YAP– stifening is a common feature of most solid tumor [148]. TEAD complex, inhibitors of YAP interaction with other Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 10 of 14

Normal ECM composition Normal ECM composition Normal ECM stiffness Normal stiffness YAP/TAZ non-active CAFs with active YAP/TAZ modification induced by cancer-associated factors

ECM remodeling Fibronectin-rich ECM and stiffening Stiff ECM Therapy-resistant cancer Aggressive phenotype

Normal CAFs with active YAP/TAZ

Therapy-sensitive Therapy-resistant cancer cancer cells cells with active YAP/TAZ

Non-fibronectin ECM fibril Fibronectin fibril

Fig. 6 A hypothetical model of CAF-mediated YAP activation in cancer cells via ECM remodeling. On the frst step soluble factors secreted by cancer cells activate YAP in CAFs, leading to ECM remodeling and stifening. Stif and fbronectin-rich ECM would promote YAP activation in both CAFs and cancer cells, which confers therapy resistance of cancer

transcriptional factors are not known. However, as was Conclusion mentioned above, YAP/TAZ-mediated transcriptional Obviously, activity of YAP/TAZ is regulated by multi- complexes independent of TEAD (such as the β-catenin– ple mechanisms. It is also worthy to note that each cell YAP–TBX5 transcriptional complex) might also drive sufers multiple mechanical and chemical stimuli from the cancer progression and therapy resistance. Tere- diferent origins. Terefore, multiple signaling cascades fore, identifcation of inhibitors of TEAD-independent initiated by individual inputs may act in parallel, syner- YAP/TAZ transcriptional complexes would contribute to gize to achieve stronger responses, or antagonize each development of novel therapeutic strategies for cancer. other with prevailing of one of the mechanisms. Such Although pro-oncogenic functions of YAP/TAZ have pleiotropic regulations would fne-tune the YAP/TAZ been revealed by numerous researches, it is worth not- transcriptional activity, enabling to control diverse bio- ing that there are several reports suggesting that YAP logical functions of YAP/TAZ. can function as a tumor suppressor and is required for Te important roles of YAP/TAZ in cell cycle progres- -mediated and cisplatin-induced cell death [18, 155, sion, tissue growth and homeostasis make these pro- 156]. For development of YAP/TAZ-targeted thera- teins potential targets for clinical application, and one peutic strategies, further investigations are needed to of the most straightforward application would be can- reveal YAP/TAZ functions in diferent cancer types. cer treatment. Given that YAP/TAZ are regulated by Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 11 of 14

multiple ways, better understanding of upstream regula- 7. Juan WC, Hong W (2016) Targeting the for tissue regeneration and cancer therapy. Genes 7:E55 tors for YAP/TAZ is essential for developing therapeutic 8. Kanai F, Marignani PA, Sarbassova D et al (2000) TAZ: a novel transcrip- approaches that target YAP/TAZ signaling. Furthermore, tional co-activator regulated by interactions with 14-3-3 and PDZ as transcriptional co-activators, YAP/TAZ interact with domain proteins. EMBO J 19:6778–6791 9. Uhlén M, Fagerberg L, Hallström BM et al (2015) Proteomics. Tissue- various transcription factors to induce expression of a based map of the human proteome. Science 347:1260419 wide range of genes, some of which drive disease pro- 10. Uhlen M, Oksvold P, Fagerberg L et al (2010) Towards a knowledge- gression, and others of which might have protective based human protein atlas. Nat Biotechnol 28:1248–1250 11. Huang J, Wu S, Barrera J et al (2005) The Hippo signaling pathway coor- efects against pathogenesis. Terefore, further studies dinately regulates cell proliferation and apoptosis by inactivating Yorkie, aiming at exhaustive investigations on YAP/TAZ down- the Drosophila homolog of YAP. Cell 122:421–434 stream efectors would be required for creating therapeu- 12. Kim M-K, Jang J-W, Bae S-C (2018) DNA binding partners of YAP/TAZ. BMB Rep 51:126–133 tic strategies with high selectivities and efcacies. 13. Hao J, Zhang Y, Wang Y et al (2014) Role of extracellular matrix and YAP/ TAZ in cell fate determination. Cell Signal 26:186–191 14. Zhang H, Liu C-Y, Zha Z-Y et al (2009) TEAD transcription factors medi- Abbreviations ate the function of TAZ in cell growth and epithelial–mesenchymal AJ: adherens junction; CAF: cancer-associated fbroblast; CIP: contact inhibi- transition. J Biol Chem 284:13355–13362 tion of proliferation; ECM: extracellular matrix; EMT: epithelial–mesenchymal 15. Pan D (2010) The Hippo signaling pathway in development and cancer. transition; ESC: embryonic stem cell; FA: focal adhesion; MAF: melanoma- Dev Cell 19:491–505 associated fbroblasts; MSC: mesenchymal stem cell; MW: molecular weight; 16. Zhao B, Li L, Lei Q, Guan K-L (2010) The Hippo-YAP pathway in organ NES: nuclear export signal; NSCLC: non-small cell lung cancer; PP2A: protein size control and tumorigenesis: an updated version. Genes Dev phosphatase 2A; RhoGAP: Rho GTPase activating protein; RhoGEF: Rho gua- 24:862–874 nine exchange factor; TAZ: WW domain containing transcription regulator 1; 17. Dupont S, Morsut L, Aragona M et al (2011) Role of YAP/TAZ in mecha- TEAD: TEA domain; VP: verteporfn; YAP: Yes-associated protein. notransduction. Nature 474:179–183 18. Guo L, Teng L (2015) YAP/TAZ for cancer therapy: opportunities and Authors’ contributions challenges (review). Int J Oncol 46:1444–1452 OD and HH wrote the manuscript. MSa and MSo reviewed and edited the 19. Gafney CJ, Oka T, Mazack V et al (2012) Identifcation, basic characteri- manuscript. All authors read and approved the fnal manuscript. zation and evolutionary analysis of diferentially spliced mRNA isoforms of human YAP1 gene. Gene 509:215–222 Author details 20. Chan EHY, Nousiainen M, Chalamalasetty RB et al (2005) The Ste20-like 1 2 R-Pharm Japan, Tokyo 105‑0001, Japan. Mechanobiology Laboratory, kinase Mst2 activates the human large tumor suppressor kinase Lats1. Nagoya University Graduate School of Medicine, 65 Tsurumai, Showa‑ku, 24:2076–2086 3 Nagoya 466‑8550, Japan. R-Pharm, Moscow 123317, Russia. 21. Zhao B, Wei X, Li W et al (2007) Inactivation of YAP oncoprotein by the Hippo pathway is involved in cell contact inhibition and tissue growth Acknowledgements control. Genes Dev 21:2747–2761 We would like to thank all the members of Mechanobiology Laboratory for 22. Callus BA, Verhagen AM, Vaux DL (2006) Association of mammalian their helpful discussions. sterile twenty kinases, Mst1 and Mst2, with hSalvador via C-terminal coiled–coil domains, leads to its stabilization and phosphorylation. Competing interests FEBS J 273:4264–4276 The authors declare that they have no competing interests. 23. Praskova M, Xia F, Avruch J (2008) MOBKL1A/MOBKL1B phosphorylation by MST1 and MST2 inhibits cell proliferation. Curr Biol 18:311–321 Funding 24. Hoa L, Kulaberoglu Y, Gundogdu R et al (2016) The characterisation of This work was supported by the grant for collaborative research between LATS2 kinase regulation in Hippo-YAP signalling. Cell Signal 28:488–497 Nagoya University and R-Pharm (2614Dj-02b). 25. Meng Z, Moroishi T, Guan K-L (2016) Mechanisms of Hippo pathway regulation. Genes Dev 30:1–17 Publisher’s Note 26. Yin F, Yu J, Zheng Y et al (2013) Spatial organization of Hippo signaling at the plasma membrane mediated by the tumor suppressor Merlin/ Springer Nature remains neutral with regard to jurisdictional claims in pub- NF2. Cell 154:1342–1355 lished maps and institutional afliations. 27. Schlegelmilch K, Mohseni M, Kirak O et al (2011) Yap1 acts downstream of α-catenin to control epidermal proliferation. Cell 144:782–795 Received: 29 April 2018 Accepted: 6 July 2018 28. Kim N-G, Koh E, Chen X, Gumbiner BM (2011) E-cadherin mediates contact inhibition of proliferation through Hippo signaling-pathway components. Proc Natl Acad Sci USA 108:11930–11935 29. Meng Z, Moroishi T, Mottier-Pavie V et al (2015) MAP4K family kinases act in parallel to MST1/2 to activate LATS1/2 in the Hippo pathway. Nat References Commun 6:8357 1. Panciera T, Azzolin L, Cordenonsi M, Piccolo S (2017) Mechanobiol- 30. Silvis MR, Kreger BT, Lien W-H, et al (2011) α-catenin is a tumor suppres- ogy of YAP and TAZ in physiology and disease. Nat Rev Mol Cell Biol sor that controls cell accumulation by regulating the localization and 18:758–770 activity of the transcriptional coactivator Yap1. Sci Signal 4:ra33 2. Fu V, Ploufe SW, Guan K-L (2017) The Hippo pathway in organ develop- 31. Aragona M, Panciera T, Manfrin A et al (2013) A mechanical checkpoint ment, homeostasis, and regeneration. Curr Opin Cell Biol 49:99–107 controls multicellular growth through YAP/TAZ regulation by actin- 3. Wang Y, Yu A, Yu F-X (2017) The Hippo pathway in tissue homeostasis processing factors. Cell 154:1047–1059 and regeneration. Protein Cell 8:349–359 32. Hirata H, Samsonov M, Sokabe M (2017) Actomyosin contractility 4. Yu F-X, Meng Z, Ploufe SW, Guan K-L (2015) Hippo pathway regulation provokes contact inhibition in E-cadherin-ligated keratinocytes. Sci Rep of gastrointestinal tissues. Annu Rev Physiol 77:201–227 7:46326 5. Yu F-X, Zhao B, Guan K-L (2015) Hippo pathway in organ size control, 33. Wood MN, Ishiyama N, Singaram I et al (2017) α-Catenin homodimers tissue homeostasis, and cancer. Cell 163:811–828 are recruited to phosphoinositide-activated membranes to promote 6. Yin M-X, Zhang L (2015) Hippo signaling in epithelial stem cells. Acta adhesion. J Cell Biol 216:3767–3783 Biochim Biophys Sin 47:39–45 Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 12 of 14

34. Poon CLC, Lin JI, Zhang X, Harvey KF (2011) The sterile 20-like kinase 60. Shiu J-Y, Aires L, Lin Z, Vogel V (2018) Nanopillar force measurements Tao-1 controls tissue growth by regulating the Salvador-Warts-Hippo reveal actin-cap-mediated YAP mechanotransduction. Nat Cell Biol pathway. Dev Cell 21:896–906 20:262–271 35. Boggiano JC, Vanderzalm PJ, Fehon RG (2011) Tao-1 phosphorylates 61. Frey S, Görlich D (2007) A saturated FG-repeat hydrogel can repro- Hippo/MST kinases to regulate the Hippo-Salvador-Warts tumor sup- duce the permeability properties of nuclear pore complexes. Cell pressor pathway. Dev Cell 21:888–895 130:512–523 36. Yu F-X, Zhao B, Panupinthu N et al (2012) Regulation of the Hippo-YAP 62. Jamali T, Jamali Y, Mehrbod M, Mofrad MRK (2011) Nuclear pore com- pathway by G-protein-coupled signaling. Cell 150:780–791 plex: biochemistry and biophysics of nucleocytoplasmic transport in 37. Wilson KE, Li Y-W, Yang N et al (2014) PTPN14 forms a complex with health and disease. Int Rev Cell Mol Biol 287:233–286 Kibra and LATS1 proteins and negatively regulates the YAP oncogenic 63. Lin KC, Moroishi T, Meng Z et al (2017) Regulation of Hippo pathway function. J Biol Chem 289:23693–23700 TEAD by p38 MAPK-induced cytoplasmic transloca- 38. Mo J-S, Meng Z, Kim YC et al (2015) Cellular energy stress induces tion. Nat Cell Biol 19:996–1002 AMPK-mediated regulation of YAP and the Hippo pathway. Nat Cell Biol 64. Furukawa KT, Yamashita K, Sakurai N, Ohno S (2017) The epithelial cir- 17:500–510 cumferential actin belt regulates YAP/TAZ through nucleocytoplasmic 39. Wang W, Xiao Z-D, Li X et al (2015) AMPK modulates Hippo pathway shuttling of merlin. Cell Rep 20:1435–1447 activity to regulate energy homeostasis. Nat Cell Biol 17:490–499 65. Wang S, Lu Y, Yin M-X et al (2016) Importin α1 mediates yorkie nuclear 40. DeRan M, Yang J, Shen C-H et al (2014) Energy stress regulates hippo- import via an N-terminal non-canonical nuclear localization signal. J YAP signaling involving AMPK-mediated regulation of -like Biol Chem 291:7926–7937 1 protein. Cell Rep 9:495–503 66. Santucci M, Vignudelli T, Ferrari S et al (2015) The Hippo pathway and 41. Reddy BVVG, Irvine KD (2013) Regulation of Hippo signaling by EGFR- YAP/TAZ–TEAD protein–protein interaction as targets for regenerative MAPK signaling through Ajuba family proteins. Dev Cell 24:459–471 medicine and cancer treatment. J Med Chem 58:4857–4873 42. Karaman R, Halder G (2017) Cell Junctions in Hippo Signaling. Cold 67. Ota M, Sasaki H (2008) Mammalian TEAD proteins regulate cell prolif- Spring Harb Perspect Biol 10:a028753 eration and contact inhibition as transcriptional mediators of Hippo 43. Burridge K, Fath K, Kelly T et al (1988) Focal adhesions: transmembrane signaling. Development 135:4059–4069 junctions between the extracellular matrix and the cytoskeleton. Annu 68. Choi H-J, Zhang H, Park H et al (2015) Yes-associated protein regulates Rev Cell Biol 4:487–525 endothelial cell contact-mediated expression of angiopoietin-2. Nat 44. Burridge K (2017) Focal adhesions: a personal perspective on a half Commun 6:6943 century of progress. FEBS J 284:3355–3361 69. Herr KJ, Tsang Y-HN, Ong JWE et al (2014) Loss of α-catenin elicits a 45. Hirata H, Sokabe M, Lim CT (2014) Molecular mechanisms underlying cholestatic response and impairs liver regeneration. Sci Rep 4:6835 the force-dependent regulation of actin-to-ECM linkage at the focal 70. Bertocchi C, Wang Y, Ravasio A et al (2017) Nanoscale architecture of adhesions. Prog Mol Biol Transl Sci 126:135–154 cadherin-based cell adhesions. Nat Cell Biol 19:28–37 46. Pelham RJ Jr, Wang YL (1997) Cell locomotion and focal adhe- 71. Mège RM, Ishiyama N (2017) Integration of cadherin adhesion and sions are regulated by substrate fexibility. Proc Natl Acad Sci USA cytoskeleton at junctions. Cold Spring Harb Perspect Biol 9:a028738 94:13661–13665 72. Liu Z, Tan JL, Cohen DM et al (2010) Mechanical tugging force regulates 47. Choquet D, Felsenfeld DP, Sheetz MP (1997) Extracellular matrix rigidity the size of cell–cell junctions. Proc Natl Acad Sci USA 107:9944–9949 causes strengthening of integrin–cytoskeleton linkages. Cell 88:39–48 73. Gladden AB, Hebert AM, Schneeberger EE, McClatchey AI (2010) The 48. Totaro A, Castellan M, Battilana G et al (2017) YAP/TAZ link cell mechan- NF2 tumor suppressor, Merlin, regulates epidermal development ics to Notch signalling to control epidermal stem cell fate. Nat Com- through the establishment of a junctional polarity complex. Dev Cell mun 8:15206 19:727–739 49. Kim J-H, Asthagiri AR (2011) Matrix stifening sensitizes epithelial cells to 74. Yonemura S, Wada Y, Watanabe T et al (2010) alpha-Catenin as a tension EGF and enables the loss of contact inhibition of proliferation. J Cell Sci transducer that induces adherens junction development. Nat Cell Biol 124:1280–1287 12:533–542 50. Engler AJ, Sen S, Sweeney HL, Discher DE (2006) Matrix elasticity directs 75. Yao M, Qiu W, Liu R et al (2014) Force-dependent conformational switch stem cell lineage specifcation. Cell 126:677–689 of α-catenin controls vinculin binding. Nat Commun 5:4525 51. Yeung T, Georges PC, Flanagan LA et al (2005) Efects of substrate 76. Zhang N, Bai H, David KK et al (2010) The Merlin/NF2 tumor suppressor stifness on cell morphology, cytoskeletal structure, and adhesion. Cell functions through the YAP oncoprotein to regulate tissue homeostasis Motil Cytoskeleton 60:24–34 in mammals. Dev Cell 19:27–38 52. Hirata H, Chiam K-H, Lim CT, Sokabe M (2014) Actin fow and talin 77. Benhamouche S, Curto M, Saotome I et al (2010) Nf2/Merlin controls dynamics govern rigidity sensing in actin-integrin linkage through talin progenitor homeostasis and tumorigenesis in the liver. Genes Dev extension. J R Soc Interface 11:20140734 24:1718–1730 53. Wada K-I, Itoga K, Okano T et al (2011) Hippo pathway regulation by cell 78. Azzolin L, Panciera T, Soligo S et al (2014) YAP/TAZ incorporation in the morphology and stress fbers. Development 138:3907–3914 β-catenin destruction complex orchestrates the Wnt response. Cell 54. Kim N-G, Gumbiner BM (2015) Adhesion to fbronectin regulates Hippo 158:157–170 signaling via the FAK–Src–PI3K pathway. J Cell Biol 210:503–515 79. Dutta S, Mana-Capelli S, Paramasivam M et al (2018) TRIP6 inhibits 55. Lachowski D, Cortes E, Robinson B et al (2018) FAK controls the Hippo signaling in response to tension at adherens junctions. EMBO mechanical activation of YAP, a transcriptional regulator required for Rep 19:337–350 durotaxis. FASEB J 32:1099–1107 80. Pasapera AM, Schneider IC, Rericha E et al (2010) Myosin II activity regu- 56. Elbediwy A, Vincent-Mistiaen ZI, Spencer-Dene B et al (2016) Integrin lates vinculin recruitment to focal adhesions through FAK-mediated signalling regulates YAP and TAZ to control skin homeostasis. Develop- paxillin phosphorylation. J Cell Biol 188:877–890 ment 143:1674–1687 81. Schimizzi GV, Longmore GD (2015) Ajuba proteins. Curr Biol 57. Sabra H, Brunner M, Mandati V et al (2017) β1 integrin-dependent Rac/ 25:R445–R446 group I PAK signaling mediates YAP activation of Yes-associated protein 82. Das Thakur M, Feng Y, Jagannathan R et al (2010) Ajuba LIM proteins 1 (YAP1) via NF2/merlin. J Biol Chem 292:19179–19197 are negative regulators of the Hippo signaling pathway. Curr Biol 58. Zhao B, Li L, Wang L et al (2012) Cell detachment activates the Hippo 20:657–662 pathway via cytoskeleton reorganization to induce anoikis. Genes Dev 83. Sun G, Irvine KD (2013) Ajuba family proteins link JNK to Hippo signal- 26:54–68 ing. Sci Signal 6:ra81 59. Elosegui-Artola A, Andreu I, Beedle AEM et al (2017) Force triggers 84. Codelia VA, Sun G, Irvine KD (2014) Regulation of YAP by mechanical YAP nuclear entry by regulating transport across nuclear pores. Cell strain through Jnk and Hippo signaling. Curr Biol 24:2012–2017 171:1397–1410 85. Ibar C, Kirichenko E, Keepers B et al (2018) Tension-dependent regulation of mammalian Hippo signaling through LIMD1. J Cell Sci 131:jcs214700 Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 13 of 14

86. Li P, Silvis MR, Honaker Y et al (2016) αE-catenin inhibits a Src–YAP1 112. Yang S, Zhang L, Purohit V et al (2015) Active YAP promotes pan- oncogenic module that couples tyrosine kinases and the efector of creatic cancer cell motility, invasion and tumorigenesis in a mitotic Hippo signaling pathway. Genes Dev 30:798–811 phosphorylation-dependent manner through LPAR3. Oncotarget 87. Lin C, Yao E, Zhang K et al (2017) YAP is essential for mechanical force 6:36019–36031 production and epithelial cell proliferation during lung branching 113. Varzavand A, Hacker W, Ma D et al (2016) α3β1 integrin suppresses morphogenesis. Elife 6:e21130 prostate cancer metastasis via regulation of the Hippo pathway. 88. Porazinski S, Wang H, Asaoka Y et al (2015) YAP is essential for tissue Cancer Res 76:6577–6587 tension to ensure vertebrate 3D body shape. Nature 521:217–221 114. Wang L, Shi S, Guo Z et al (2013) Overexpression of YAP and TAZ is an 89. Calvo F, Ege N, Grande-Garcia A et al (2013) Mechanotransduction and independent predictor of prognosis in and related YAP-dependent matrix remodelling is required for the generation and to the proliferation and metastasis of colon cancer cells. PLoS ONE maintenance of cancer-associated fbroblasts. Nat Cell Biol 15:637–646 8:e65539 90. Bai H, Zhu Q, Surcel A et al (2016) Yes-associated protein impacts 115. Yuan Y, Li D, Li H et al (2016) YAP overexpression promotes the epi- adherens junction assembly through regulating actin cytoskeleton thelial–mesenchymal transition and chemoresistance in pancreatic organization. Am J Physiol Gastrointest Liver Physiol 311:G396–G411 cancer cells. Mol Med Rep 13:237–242 91. Qiao Y, Chen J, Lim YB et al (2017) YAP regulates actin dynamics through 116. Liu J-Y, Li Y-H, Lin H-X et al (2013) Overexpression of YAP 1 contrib- ARHGAP29 and promotes metastasis. Cell Rep 19:1495–1502 utes to progressive features and poor prognosis of human urothelial 92. Dupont S (2016) Role of YAP/TAZ in cell-matrix adhesion-mediated carcinoma of the bladder. BMC Cancer 13:349 signalling and mechanotransduction. Exp Cell Res 343:42–53 117. Lau AN, Curtis SJ, Fillmore CM et al (2014) Tumor-propagating cells 93. Chang C, Goel HL, Gao H et al (2015) A laminin 511 matrix is regulated and Yap/Taz activity contribute to lung tumor progression and by TAZ and functions as the for the α6Bβ1 integrin to sustain metastasis. EMBO J 33:468–481 stem cells. Genes Dev 29:1–6 118. Bartucci M, Dattilo R, Moriconi C et al (2015) TAZ is required for 94. Shao DD, Xue W, Krall EB et al (2014) KRAS and YAP1 converge to regu- metastatic activity and chemoresistance of breast cancer stem cells. late EMT and tumor survival. Cell 158:171–184 Oncogene 34:681–690 95. Chen Q, Zhang N, Gray RS et al (2014) A temporal requirement for 119. Rosenbluh J, Nijhawan D, Cox AG et al (2012) β-Catenin-driven Hippo signaling in mammary gland diferentiation, growth, and tumori- cancers require a YAP1 transcriptional complex for survival and genesis. Genes Dev 28:432–437 tumorigenesis. Cell 151:1457–1473 96. Cai J, Zhang N, Zheng Y et al (2010) The Hippo signaling pathway 120. Li S, Yu Z, Chen S-S et al (2015) The YAP1 oncogene contributes to blad- restricts the oncogenic potential of an intestinal regeneration pro- der cancer cell proliferation and migration by regulating the H19 long gram. Genes Dev 24:2383–2388 noncoding RNA. Urol Oncol 33(427):e1–e10 97. Hossain Z, Ali SM, Ko HL et al (2007) Glomerulocystic kidney disease 121. Ciamporcero E, Shen H, Ramakrishnan S et al (2016) YAP activation pro- in mice with a targeted inactivation of Wwtr1. Proc Natl Acad Sci USA tects urothelial cell carcinoma from treatment-induced DNA damage. 104:1631–1636 Oncogene 35:1541–1553 98. Reginensi A, Scott RP, Gregorief A et al (2013) Yap- and Cdc42- 122. Bi L, Ma F, Tian R et al (2018) AJUBA increases the cisplatin resistance dependent nephrogenesis and morphogenesis during mouse kidney through hippo pathway in cervical cancer. Gene 644:148–154 development. PLoS Genet 9:e1003380 123. Song S, Honjo S, Jin J et al (2015) The Hippo coactivator YAP1 mediates 99. Grijalva JL, Huizenga M, Mueller K et al (2014) Dynamic alterations in EGFR overexpression and confers chemoresistance in Esophageal Hippo signaling pathway and YAP activation during liver regenera- cancer. Clin Cancer Res 21:2580–2590 tion. Am J Physiol Gastrointest Liver Physiol 307:G196–G204 124. Cheng H, Zhang Z, Rodriguez-Barrueco R et al (2016) Functional 100. Yimlamai D, Christodoulou C, Galli GG et al (2014) Hippo pathway genomics screen identifes YAP1 as a key determinant to enhance treat- activity infuences liver cell fate. Cell 157:1324–1338 ment sensitivity in lung cancer cells. Oncotarget 7:28976–28988 101. Ohgushi M, Minaguchi M, Sasai Y (2015) Rho-signaling-directed 125. Zhao Y, Yang X (2015) The Hippo pathway in chemotherapeutic drug YAP/TAZ activity underlies the long-term survival and expansion of resistance. Int J Cancer 137:2767–2773 human embryonic stem cells. Cell Stem Cell 17:448–461 126. Fernandez-L A, Squatrito M, Northcott P et al (2012) Oncogenic YAP 102. Musah S, Wrighton PJ, Zaltsman Y et al (2014) Substratum-induced promotes radioresistance and genomic instability in medulloblastoma diferentiation of human pluripotent stem cells reveals the coactiva- through IGF2-mediated Akt activation. Oncogene 31:1923–1937 tor YAP is a potent regulator of neuronal specifcation. Proc Natl Acad 127. Lin L, Sabnis AJ, Chan E et al (2015) The Hippo efector YAP promotes Sci USA 111:13805–13810 resistance to RAF- and MEK-targeted cancer therapies. Nat Genet 103. Beyer TA, Weiss A, Khomchuk Y et al (2013) Switch enhancers 47:250–256 interpret TGF-β and Hippo signaling to control cell fate in human 128. Kim MH, Kim J, Hong H et al (2016) Actin remodeling confers BRAF embryonic stem cells. Cell Rep 5:1611–1624 inhibitor resistance to melanoma cells through YAP/TAZ activation. 104. Tang Y, Rowe RG, Botvinick EL et al (2013) MT1-MMP-dependent EMBO J 35:462–478 control of skeletal stem cell commitment via a β1-integrin/YAP/TAZ 129. Cottini F, Hideshima T, Xu C et al (2014) Rescue of Hippo coactiva- signaling axis. Dev Cell 25:402–416 tor YAP1 triggers DNA damage-induced apoptosis in hematological 105. Zhong W, Tian K, Zheng X et al (2013) Mesenchymal stem cell and cancers. Nat Med 20:599–606 chondrocyte fates in a multishear microdevice are regulated by Yes- 130. Barry ER, Morikawa T, Butler BL et al (2013) Restriction of intestinal associated protein. Stem Cells Dev 22:2083–2093 stem cell expansion and the regenerative response by YAP. Nature 106. Cosgrove BD, Mui KL, Driscoll TP et al (2016) N-cadherin adhesive 493:106–110 interactions modulate matrix mechanosensing and fate commitment 131. Bhat KPL, Salazar KL, Balasubramaniyan V et al (2011) The transcriptional of mesenchymal stem cells. Nat Mater 15:1297–1306 coactivator TAZ regulates mesenchymal diferentiation in malignant 107. Zanconato F, Cordenonsi M, Piccolo S (2016) YAP/TAZ at the roots of glioma. Genes Dev 25:2594–2609 cancer. Cancer Cell 29:783–803 132. Dong J, Feldmann G, Huang J et al (2007) Elucidation of a universal size- 108. Zanconato F, Piccolo S (2016) Eradicating tumor drug resistance at its control mechanism in Drosophila and mammals. Cell 130:1120–1133 YAP-biomechanical roots. EMBO J 35:459–461 133. Harvey KF, Zhang X, Thomas DM (2013) The Hippo pathway and human 109. Cao J, Huang W (2017) Two faces of Hippo: activate or suppress the cancer. Nat Rev Cancer 13:246–257 Hippo pathway in cancer. Anticancer Drugs 28:1079–1085 134. Tao J, Calvisi DF, Ranganathan S et al (2014) Activation of β-catenin and 110. Shah SR, David JM, Tippens ND et al (2017) Brachyury-YAP regulatory Yap1 in human hepatoblastoma and induction of hepatocarcinogen- axis drives stemness and growth in cancer. Cell Rep 21:495–507 esis in mice. Gastroenterology 147:690–701 111. Hua G, Lv X, He C et al (2016) YAP induces high-grade serous 135. Xia H, Dai X, Yu H et al (2018) EGFR-PI3 K-PDK1 pathway regulates YAP carcinoma in fallopian tube secretory epithelial cells. Oncogene signaling in hepatocellular carcinoma: the mechanism and its implica- 35:2247–2265 tions in targeted therapy. Cell Death Dis 9:269 Dobrokhotov et al. Clin Trans Med (2018) 7:23 Page 14 of 14

136. Ghiso E, Migliore C, Ciciriello V et al (2017) YAP-dependent AXL over- 147. Tamm C, Böwer N, Annerén C (2011) Regulation of mouse embryonic expression mediates resistance to EGFR inhibitors in NSCLC. Neoplasia stem cell self-renewal by a Yes–YAP–TEAD2 signaling pathway down- 19:1012–1021 stream of LIF. J Cell Sci 124:1136–1144 137. Jerhammar F, Johansson A-C, Ceder R et al (2014) YAP1 is a potential 148. Paszek MJ, Zahir N, Johnson KR et al (2005) Tensional homeostasis and biomarker for cetuximab resistance in head and neck cancer. Oral the malignant phenotype. Cancer Cell 8:241–254 Oncol 50:832–839 149. Ishihara S, Inman DR, Li W-J et al (2017) Mechano- 138. Lee JE, Park HS, Lee D et al (2016) Hippo pathway efector YAP inhibition in mesenchymal stem cells induces prosaposin secretion to drive the restores the sensitivity of EGFR-TKI in lung adenocarcinoma having pri- proliferation of breast cancer cells. Cancer Res 77:6179–6189 mary or acquired EGFR-TKI resistance. Biochem Biophys Res Commun 150. Liu-Chittenden Y, Huang B, Shim JS et al (2012) Genetic and pharmaco- 474:154–160 logical disruption of the TEAD–YAP complex suppresses the oncogenic 139. Lee BS, Park DI, Lee DH et al (2017) Hippo efector YAP directly regulates activity of YAP. Genes Dev 26:1300–1305 the expression of PD-L1 transcripts in EGFR-TKI-resistant lung adenocar- 151. Fisher ML, Grun D, Adhikary G et al (2017) Inhibition of YAP function cinoma. Biochem Biophys Res Commun 491:493–499 overcomes BRAF inhibitor resistance in melanoma cancer stem cells. 140. Xu MZ, Chan SW, Liu AM et al (2011) AXL receptor kinase is a mediator Oncotarget 8:110257–110272 of YAP-dependent oncogenic functions in hepatocellular carcinoma. 152. Zhou Z, Hu T, Xu Z et al (2015) Targeting Hippo pathway by specifc Oncogene 30:1229–1240 interruption of YAP–TEAD interaction using cyclic YAP-like peptides. 141. Hsu P-C, You B, Yang Y-L et al (2016) YAP promotes erlotinib resistance FASEB J 29:724–732 in human non-small cell lung cancer cells. Oncotarget 7:51922–51933 153. Jiao S, Wang H, Shi Z et al (2014) A peptide mimicking VGLL4 function 142. Cross SE, Jin Y-S, Rao J, Gimzewski JK (2007) Nanomechanical analysis of acts as a YAP antagonist therapy against gastric cancer. Cancer Cell cells from cancer patients. Nat Nanotechnol 2:780–783 25:166–180 143. Arpin M, Chirivino D, Naba A, Zwaenepoel I (2011) Emerging role for 154. Pobbati AV, Han X, Hung AW et al (2015) Targeting the central pocket ERM proteins in cell adhesion and migration. Cell Adh Migr 5:199–206 in human transcription factor TEAD as a potential cancer therapeutic 144. Kuzet S-E, Gaggioli C (2016) Fibroblast activation in cancer: when seed strategy. Structure 23:2076–2086 fertilizes soil. Cell Tissue Res 365:607–619 155. Danovi SA, Rossi M, Gudmundsdottir K et al (2008) Yes-associated 145. Erdogan B, Webb DJ (2017) Cancer-associated fbroblasts modulate protein (YAP) is a critical mediator of c-Jun-dependent apoptosis. Cell signaling and extracellular matrix remodeling to regulate Death Difer 15:217–219 tumor metastasis. Biochem Soc Trans 45:229–236 156. Strano S, Munarriz E, Rossi M et al (2001) Physical interaction with Yes- 146. Hirata E, Girotti MR, Viros A et al (2015) Intravital imaging reveals how associated protein enhances p73 transcriptional activity. J Biol Chem BRAF inhibition generates drug-tolerant microenvironments with high 276:15164–15173 integrin β1/FAK signaling. Cancer Cell 27:574–588